Advancing drug–target interaction prediction: a comprehensive graph-based approach integrating knowledge graph embedding and ProtBert pretraining

Djeddi, Warith Eddine; Hermi, Khalil; Ben Yahia, Sadok; Diallo, Gayo

doi:10.1186/s12859-023-05593-6

Table 4 Performance comparison of DTIOG variants on the ENZ dataset using KGE

From: Advancing drug–target interaction prediction: a comprehensive graph-based approach integrating knowledge graph embedding and ProtBert pretraining

Approaches	AUPR	AUC	ACC	MCC	F1
Cosine similarity
DTIOG-RF	0.963	0.994	0.991	0.958	0.962
DTIOG-DT	0.976	0.996	0.994	0.972	0.975
DTIOG-MLP	0.973	0.996	0.980	0.902	0.913
DTIOG-KNeighbors	0.973	0.996	0.968	0.869	0.879
DTIOG-BaggingClassifier	0.986	0.998	0.970	0.878	0.887
DTIOG-GradientBoosting	0.509	0.836	0.902	0.416	0.436
DTIOG-GaussianNB	0.586	0.840	0.848	0.452	0.513
DTIOG-SGD	0.661	0.922	0.921	0.574	0.608
DTIOG-ETC	0.967	0.995	0.991	0.962	0.966
Euclidean distance
DTIOG-RF	0.964	0.995	0.991	0.959	0.963
DTIOG-DT	0.976	0.996	0.994	0.973	0.976
DTIOG-MLP	0.950	0.992	0.970	0.854	0.871
DTIOG-KNeighbors	0.974	0.996	0.970	0.876	0.885
DTIOG-BaggingClassifier	0.986	0.998	0.974	0.890	0.899
DTIOG-GradientBoosting	0.459	0.808	0.902	0.418	0.434
DTIOG-GaussianNB	0.583	0.833	0.831	0.423	0.486
DTIOG-SGD	0.656	0.922	0.919	0.561	0.595
DTIOG-ETC	0.967	0.995	0.992	0.963	0.966
Manhattan distance
DTIOG-RF	0.965	0.995	0.991	0.960	0.964
DTIOG-DT	0.976	0.996	0.994	0.972	0.975
DTIOG-MLP	0.952	0.993	0.970	0.855	0.872
DTIOG-KNeighbors	0.981	0.997	0.974	0.893	0.902
DTIOG-BaggingClassifier	0.989	0.998	0.976	0.898	0.907
DTIOG-GradientBoosting	0.612	0.885	0.915	0.526	0.553
DTIOG-GaussianNB	0.584	0.838	0.838	0.440	0.502
DTIOG-SGD	0.662	0.923	0.917	0.554	0.585
DTIOG-ETC	0.969	0.995	0.992	0.964	0.968
Jaccard similarity
DTIOG-RF	0.962	0.994	0.990	0.956	0.960
DTIOG-DT	0.975	0.996	0.993	0.971	0.974
DTIOG-MLP	0.956	0.993	0.971	0.855	0.870
DTIOG-KNeighbors	0.963	0.994	0.960	0.844	0.855
DTIOG-BaggingClassifier	0.979	0.997	0.963	0.852	0.862
DTIOG-GradientBoosting	0.606	0.879	0.914	0.512	0.539
DTIOG-GaussianNB	0.569	0.827	0.774	0.371	0.432
DTIOG-SGD	0.658	0.921	0.921	0.570	0.603
DTIOG-ETC	0.966	0.995	0.991	0.961	0.965
Pearson correlation coefficient
DTIOG-RF	0.963	0.994	0.991	0.958	0.962
DTIOG-DT	0.975	0.996	0.994	0.971	0.974
DTIOG-MLP	0.973	0.996	0.980	0.906	0.917
DTIOG-KNeighbors	0.970	0.995	0.962	0.851	0.861
DTIOG-BaggingClassifier	0.985	0.998	0.965	0.860	0.869
DTIOG-GradientBoosting	0.650	0.900	0.920	0.556	0.583
DTIOG-GaussianNB	0.578	0.843	0.855	0.456	0.519
DTIOG-SGD	0.656	0.923	0.919	0.560	0.594
DTIOG-ETC	0.967	0.995	0.991	0.962	0.966

The best results are indicated in bold concerning each row

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com

BMC Bioinformatics

Contact us